Gemini AI Forward Deployed Engineer
About The Position
Capgemini is seeking a deeply technical Gemini AI Forward Deployed Engineer to design, code, deploy, and troubleshoot enterprise agentic AI solutions using Gemini and Vertex AI on Google Cloud. This is a hands-on engineering role, rather than primarily an AI advisory or solution-architecture position. The engineer will take incomplete business problems through to working code, build agents and integrations, deploy solutions, evaluate their behavior, and debug failures in customer environments. The role involves collaborating with cross-functional teams to deliver technology solutions that create business value and improve client experiences. The engineer will develop scalable, high-quality solutions in a collaborative environment, with a focus on Gemini and agent engineering, retrieval-augmented generation and grounding, evaluation and reliability, and Google Cloud engineering. The position is hybrid, with work locations listed in Chicago, Atlanta, Nashville, Dallas, and New York. The posted base compensation range is $88,544 to $207,401. Capgemini offers a comprehensive benefits package to regular, full-time employees, with benefits subject to local policy and eligibility.
Responsibilities
- Build production AI applications using Gemini models and Vertex AI.
- Design and implement AI agents using Google ADK and applicable Google agent capabilities.
- Develop agent tools and function-calling integrations.
- Implement agent state, sessions, context management, and memory patterns where appropriate.
- Develop single-agent and multi-agent orchestration.
- Integrate agents with APIs, databases, enterprise applications, and operational systems.
- Implement human-in-the-loop approval and controlled-action patterns.
- Design robust error handling, retries, and fallback behaviors.
- Build enterprise RAG pipelines, including document ingestion, chunking, embeddings, indexing, retrieval, and grounding.
- Connect Gemini agents securely to enterprise knowledge and structured business data.
- Optimize retrieval quality and grounded response generation; diagnose hallucination, context, and retrieval failures.
- Build repeatable evaluation pipelines, golden datasets, and regression tests for agent behavior.
- Evaluate tool selection, groundedness, and response quality; trace agent and tool execution paths.
- Debug agent loops, incorrect tool calls, and orchestration failures.
- Implement guardrails and production observability; analyze latency, reliability, and model/runtime behavior.
- Build supporting services primarily using Python and develop REST and event-driven integrations.
- Deploy AI applications and services on Google Cloud Platform.
- Work with IAM, service accounts, secrets, and secure application patterns.
- Containerize workloads using Docker and deploy on appropriate serverless or Kubernetes platforms.
- Implement CI/CD and Terraform, as well as logging, tracing, monitoring, and operational diagnostics.
Benefits
- Paid time off, including vacation (12–25 days depending on employee grade), company-paid holidays, personal days, and sick leave.
- Medical, dental, and vision coverage (or provincial healthcare coordination in Canada).
- Retirement savings plans, such as 401(k) in the U.S. or RRSP in Canada.
- Life and disability insurance.
- Employee assistance programs.
- The role may be eligible for variable incentives, bonuses, or commissions, depending on the position and applicable laws.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Education Level
No Education Listed
